1. Field of the Invention
This invention relates to the design, construction, and use of magnetic resonance (MR) imaging to identify areas within a patient where changes in a molecular environment are occurring, as from chemical concentration changes effected by medical procedures. The invention also describes a drug delivery device for targeted drug delivery into a patient using magnetic resonance (MR) imaging combined with conventional catheter placement techniques, particularly including neurosurgical or neuroradiologic techniques used in intracranial drug delivery.
2. Background of the Art
Although endoscopic, arthroscopic, and endovascular therapies have produced significant advances in healthcare, the diagnostic accuracy and clinical utility of these procedures is ultimately "surface limited" by what the surgeon can see through the device itself or otherwise visualize during the course of the procedure. Magnetic Resonance (MR) imaging, by comparison, overcomes this limitation by enabling the surgeon to noninvasively visualize tissue planes beyond the surface of the tissue under direct evaluation. Moreover, MR imaging enables differentiation of normal from abnormal tissues, and can display critical structures such as blood vessels in three dimensions. Thus, high-speed MR-guided therapy offers an improved opportunity to maximize the benefits of minimally invasive procedures. Prototype high-speed MR imagers which permit continuous real-time visualization of tissues during surgical and endovascular procedures have already been developed. Recent publications in the medical literature have described a number of MR-guided interventions including needle biopsies, interstitial laser therapy, interstitial cryotherapy and interstitial focused ultrasound surgery.
The standard current procedure for drug treatment of various focal neurological disorders, neurovascular diseases, and neurodegenerative processes requires neurosurgeons or interventional neuroradiologists to deliver drug agents by catheters or other tubular devices directed into the cerebrovascular or cerebroventricular circulation, or by direct injection of the drug agent, or cells which biosynthesize the drug agent, into targeted intracranial tissue locations. The blood-brain barrier and blood-cerebrospinal fluid barrier almost entirely exclude large molecules like proteins, and control entry of smaller molecules. Small molecules (&lt;200 daltons) which are lipid-soluble, not bound to plasma proteins, and minimally ionized, such as nicotine, ethanol, and some chemotherapeutic agents, readily enter the brain. Water soluble molecules cross the barriers poorly or not at all. Delivery of a drug into a ventricle bypasses the blood-brain barrier, and allows for a wide distribution of the drug in the brain ventricles, cisterns, and spaces due to the normal flow pathways of cerebrospinal fluid in the brain. However, following intracerebroventricular injection, many therapeutic drug agents, particularly large-molecular weight hydrophobic drugs, fail to reach their target receptors in brain parenchyma because of metabolic inactivation and inability to diffuse into brain tissues, which may be up to 18 mm from a cerebrospinal fluid surface.
To optimize a drug's therapeutic efficacy, it should be delivered to its target tissue at the appropriate concentration. A number of studies reported in the medical literature, for example, Schmitt, Neuroscience, 13, 1984, pp. 991-1001, have shown that numerous classes of biologically active drugs, such as peptides, biogenic amines, and enkephalins have specific receptor complexes localized at particular cell regions of the brain. Placing a drug delivery device directly into brain tissue thus has the notable advantage of initially localizing the injected drug within a specific brain region containing receptors for that drug agent. Targeted drug delivery directly into tissues also reduces drug dilution, metabolism and excretion, thereby significantly improving drug efficacy, while at the same time it minimizes systemic side-effects.
An important issue in targeted drug delivery is the accuracy of the navigational process used to direct the movement of the drug delivery device. Magnetic resonance imaging will likely play an increasingly important role in optimizing drug treatment of neurological disorders. One type of MR unit designed for image-guided therapy is arranged in a "double-donut" configuration, in which the imaging coil is split axially into two components. Imaging studies are performed with this system with the surgeon standing in the axial gap of the magnet and carrying out procedures on the patient. A second type of high-speed MR imaging system combines high-resolution MR imaging with conventional X-ray fluoroscopy and digital subtraction angiography (DSA) capability in a single hybrid unit. Both of these new generations of MR scanners provide frequently updated images of the anatomical structures of interest. This real-time imaging capability makes it possible to use high-speed MR imaging to direct the movement of catheters and other drug delivery vehicles to specific tissue locations, and thereby observe the effects of specific interventional procedures.
A prerequisite for MRI-guided drug delivery into the brain parenchyma, cerebral fluid compartments, or cerebral vasculature is the availability of suitable access devices. U.S. Pat. No. 5,571,089 to Crocker et al. and U.S. Pat. No. 5,514,092 to Forman et al. disclose endovascular drug delivery and dilatation drug delivery catheters which can simultaneously dilate and deliver medication to a vascular site of stenosis. U.S. Pat. No. 5,171,217 to March describes the delivery of several specific compounds through direct injection of microcapsules or microparticles using multiple-lumen catheters, such as disclosed by Wolinsky in U.S. Pat. No. 4,824,436. U.S. Pat. No. 5,580,575 to Unger et al. discloses a method of administering drugs using gas-filled liposomes comprising a therapeutic compound, and inducing the rupture of the liposomes with ultrasound energy. U.S. Pat. No. 5,017,566 to Bodor discloses redox chemical systems for brain-targeted drug delivery of various hormones, neurotransmitters, and drugs through the intact blood-brain barrier. U.S. Pat. No. 5,226,902 to Bae et al. and U.S. Pat. No. 4,973,304 to Graham et al. disclose drug delivery devices, in which biologically active materials present within a reversibly permeable hydrogel compartment can be delivered into tissues by various endogenous and exogenous stimuli. U.S. Pat. No. 5,167,625 to Jacobsen et al. discloses an implantable drug delivery system utilizing multiple drug compartments which are activated by an electrical circuit. U.S. Pat. No. 4,941,874 to Sandow et al. discloses a device for the injection of implants, including drug implants that may used in the treatment of diseases. U.S. Pat. Nos. 4,892,538, 4,892,538, 5,106,627, 5,487,739 and 5,607,418 to Aebischer et al. disclose implantable drug therapy systems for local delivery of drugs, cells and neurotransmitters into the brain, spinal cord, and other tissues using delivery devices with a semipermeable membrane disposed at the distal end. U.S. Pat. No. 5,120,322 to Davis et al. describes the process of coating the surface layer of a stent or shunt with lathyrogenic agent to inhibit scar formation during reparative tissue formation, thereby extending exposure to the drug agent. U.S. Pat. No. 4,807,620 to Strul and 5,087,256 to Taylor are examples of catheter-based devices which convert electromagnetic Rf energy to thermal energy. Technology practiced by STS Biopolymers (Henrietta, N.Y.) allows incorporation of pharmaceutical agents into thin surface coatings during or after product manufacture. The invention disclosed by STS Biopolymers allows for the drugs to diffuse out of the coating at a controlled rate, thereby maintaining therapeutic drug levels at the coating surface while minimizing systemic concentrations. The coating can incorporate natural or synthetic materials that act as antibiotics, anticancer agents, and antithrombotics, according to the issued patent. U.S. Pat. No. 5,573,668 to Grosh et al. discloses a microporous drug delivery membrane based on an extremely thin hydrophilic shell. U.S. Pat. No. 5,569,197 to Helmus et al. discloses a drug device guidewire formed as a hollow tube suitable for drug infusion in thrombolytic and other intraluminal procedures.
A number of articles published in the medical literature, for example, Chandler et al., Ann. N.Y. Acad. Sci., 531, 1988, pp. 206-212, Bouvier et al., Neurosurgery 20(2), 1987, pp. 286-291, Johnston et al., Ann. N.Y. Acad. Sci., 531, 1988, pp. 57-67, and Sendelbeck et al., Brain Res., 328, 1985, pp. 251-258 describe implantable pump systems designed for continuous or episodic delivery of therapeutic drugs into the central nervous system via systemic, intrathecal, intracerebroventricular, and intraparenchymal injection or infusion.
The patented inventions referenced above provide useful methods for introducing, delivering, or applying a drug agent to a specific target tissue, but each invention also has inherent problems. For example, some catheter systems which provide endovascular drug delivery require temporary blocking of a segment of the vessel, thereby transiently disrupting brain perfusion. Microencapsulated coatings on catheters permit longer exposure of the tissue to the drug agent, but the physical limitations imposed by microcapsules restrict the volume and concentration of drug that can be effectively applied to any tissue area. Exposed coatings on catheters which contain drug agents usually require some type of sheath that must be removed from the catheter before the drug can be released from the coating. The sheath and any catheter components required to physically manipulate the sheath greatly increase the profile of the catheter, and thereby limit the variety of applications for which the drug delivery system can be employed. Furthermore, the binders or adhesives used in catheter coatings may themselves significantly dilute the concentration of the therapeutic agent. Finally, thermal and light energy required to melt and bond coatings such as macroaggregated albumin, to reduce tissue mass by ablation, and to inhibit restenosis by cytotoxic irradiation may also cause damage to blood vessels.
U.S. Pat. No. 5,470,307 to Lindall discloses a low-profile catheter system with an exposed coating containing a therapeutic drug agent, which can be selectively released at remote tissue site by activation of a photosensitive chemical linker. In the invention disclosed by Lindall, the linker is attached to the substrate via a complementary chemical group, which is functionalized to accept a complementary bond to the therapeutic drug agent. The drug agent is in turn bonded to a molecular lattice to accommodate a high molecular concentration per unit area and the inclusion of ancillary compounds such as markers or secondary emitters.
Although U.S. Pat. No. 5,470,307 to Lindall describes significant improvements over previous catheter-based drug delivery systems, there are nonetheless some problems. First, in common with other currently used endovascular access devices, such as catheters, microcatheters, and guidewires, the catheter tip is difficult to see on MRI because of inadequate contrast with respect to surrounding tissues and structures. This makes accurate localization difficult and degrades the quality of the diagnostic information obtained from the image. Also, the mere observation of the location of the catheter in the drug delivery system does not reliably or consistently identify the position, movement and/or efficient delivery of drugs provided through the system. Thus, one objective of this invention is to provide for an MR-compatible and visible device that significantly improves the efficacy and safety of drug delivery using MR guidance.
Any material that might be added to the structure of a pliable catheter to make it MR visible must not contribute significantly to the overall magnetic susceptibility of the catheter, or imaging artifacts could be introduced during the MR process. Moreover, forces might be applied to such a catheter by the superconducting magnetic manipulation coils of a nonlinear magnetic stereotaxis system which might be used in the practice of the present invention. In either case, the safety and efficacy of the procedure might be jeopardized, with resulting increased risk to the patient. Also, an MR-visible catheter must be made of material that is temporally stable and of low thrombolytic potential if it is to be left indwelling in either the parenchymal tissues or the cerebral vasculature. Examples of such biocompatible and MR-compatible materials which could be used to practice the invention include elastomeric hydrogel, nylon, teflon, polyamide, polyethylene, polypropylene, polysulfone, ceramics, cermets steatite, carbon fiber composites, silicon nitride, and zirconia, plexiglass, and poly-ether-ether-ketone.
It is also important that drug delivery devices used under MR guidance are MR-compatible in both static and time-varying magnetic fields. Although the mechanical effects of the magnetic field on ferromagnetic devices present the greatest danger to patients through possible unintended movement of the devices, tissue and device heating may also result from radio-frequency power deposition in electrically conductive material located within the imaging volume. Consequently, all cables, wires, and devices positioned within the MR imager must be made of materials that have properties that make them compatible with their use in human tissues during MR imaging procedures. Many materials with acceptable MR-compatibility, such as ceramics, composites and thermoplastic polymers, are electrical insulators and do not produce artifacts or safety hazards associated with applied electric fields. Some metallic materials, such as copper, brass, magnesium and aluminum are also generally MR-compatible, viz. large masses of these materials can be accommodated within the imaging region without significant image degradation.
Guidewires for the catheter or drug delivery system are usually made of radiopaque material so that their precise location can be identified during a surgical procedure through fluoroscopic viewing. Exemplary of guidewires used under X-ray viewing is the guidewire disclosed by LeVeen, U.S. Pat. No. 4,448,195, in which a radiopaque wire can be identified on fluoroscopic images by metered bands placed at predetermined locations. The U.S. Pat. No. 4,922,924, awarded to Gambale et al. discloses a bifilar arrangement whereby radiopaque and radiotransparent filaments are wrapped on a mandril to form a bifilar coil which provides radiopaque and radiotransparent areas on the guide wire. U.S. Pat. No. 5,375,596 to Twiss et al. discloses a method for locating catheters and other tubular medical devices implanted in the human body using an integrated system of wire transmitters and receivers. U.S. Pat. No. 4,572,198 to Codrington also provides for conductive elements, such as electrode wires, for systematically disturbing the magnetic field in a defined portion of a catheter to yield increased MR visibility of that region of the catheter. However, the presence of conductive elements in the catheter also introduces increased electronic noise and the possibility of Ohmic heating, and these factors have the overall effect of degrading the quality of the MR image and raising concerns about patient safety. Thus, in all of these examples of implantable medical probes, the presence of MR-incompatible wire materials causes large imaging artifacts. The lack of clinically adequate MR visibility and/or imaging artifact contamination caused by the device is also a problem for most commercially available catheters, microcatheters and shunts.
MRI enables image-guided placement of a catheter or other drug delivery device at targeted intracranial loci. High-resolution visual images denoting the actual position of the drug delivery device within the brain would be extremely useful to the clinician in maximizing the safety and efficacy of the procedure. Drug delivery devices, such as catheters, that are both MR-visible and radio-opaque could be monitored by both X-ray fluoroscopy and MR imaging, thus making intra-operative verification of catheter location possible.
Initial attempts to position and visualize endovascular devices in MR imaging were based on passive susceptibility artifacts produced by the device when exposed to the MR field. Magnetic susceptibility is a quantitative measure of a material's tendency to interact with and distort an applied magnetic field. U.S. Pat. No. 4,827,931, to Longmore and U.S. Pat. Nos. 5,154,179 and 4,989,608 to Ratner disclose the incorporation of paramagnetic material into endovascular devices to make the devices visible under MR imaging. U.S. Pat. No. 5,211,166 to Sepponen similarly discloses the use of surface impregnation of various "relaxants", including paramagnetic materials and nitrogen radicals, onto surgical instruments to enable their MR identification. However, these patents do not provide for artifact-free MR visibility in the presence of rapidly alternating magnetic fields, such as would be produced during echo-planar MR imaging pulse sequences used in real-time MR guidance of intracranial drug delivery procedures. Nor do these patents teach a method for monitoring with MR-visible catheters the outcomes of therapeutic interventions, such as, for example, drug delivery into brain tissues, cerebral ventricles, or subarachnoid space. Ultrafast imaging sequences generally have significantly lower spatial resolution than conventional spin-echo sequences. Image distortion may include general signal loss, regional signal loss, general signal enhancement, regional signal enhancement, and increased background noise. The magnetic susceptibility artifact produced by the device should be small enough not to obscure surrounding anatomy, or mask low-threshold physiological events that have an MR signature, and thereby compromise the physician's ability to perform the intervention. These relationships will be in part dependent upon the combined or comparative properties of the device, the particular drug, and the surrounding environment (e.g., tissue).
An improved method for passive MR visualization of implantable medical devices has recently been disclosed by Werne (Ser. No. 08/554,446) ITI Medical Technologies (Application Pending). This invention minimizes MR susceptibility artifacts, and controls eddy currents in the electromagnetic scattering environment, so that a bright "halo" artifact is created to outline the device in its approximately true size, shape, and position. In the method of the invention disclosed by ITI, an ultra thin coating of conductive material comprising 1-10% of the theoretical skin depth of the material being imaged--typically about 250,000 angstroms--is applied. By using a coating of 2,000-25,000 angstroms thickness, ITI has found that the susceptibility artifact due to the metal is negligible due to the low material mass. At the same time, the eddy currents are limited due to the ultra-thin conductor coating on the device. A similar method employing a nitinol wire with Teflon coat in combination with extremely thin wires of a stainless steel alloy included between the nitinol wire and Teflon coat, has recently been reported in the medical literature by Frahm et al., Proc. ISMRM, 3, 1997, p. 1931.
Exemplary of methods for active MR visualization of implanted medical devices is U.S. Pat. No. 5,211,165 to Dumoulin et al., which discloses an MR tracking system for a catheter based on transmit/receive microcoils positioned near the end of the catheter by which the position of the device can be tracked and localized. Applications of such catheter-based devices in endovascular and endoscopic imaging have been described in the medical literature, for example, Hurst et al., Mag. Res. Med., 24, 1992, pp. 343-357, Kantor et al., Circ. Res., 55, 1984, pp. 55-60; Kandarpa et al., Radiology, 181, 1991, pp. 99; Bomert et al., Proc. ISMRM, 3, 1997, p. 1925; Coutts et al., Proc. ISMRM, 3, 1997, p. 1924; Wendt et al., Proc, ISMRM, 3, 1997, p. 1926; Langsaeter et al., Proc. ISMRM, 3, 1997, p. 1929; Zimmerman et al., Proc. ISMRM, 3, 1997, p. 1930; and, Ladd et al., Proc. ISMRM, 3, 1997, p. 1937.
In the treatment of neurological diseases and disorders, targeted drug delivery can significantly improve therapeutic efficacy, while minimizing systemic side-effects of the drug therapy. Image-guided placement of the tip of a drug delivery catheter directly into specific regions of the brain can initially produce maximal drug concentration close to the loci of tissue receptors following injection of the drug. At the same time, the limited distribution of drug injected from a single catheter tip presents other problems. For example, the volume flow rate of drug delivery must be very low in order to avoid indiscriminate damage to brain cells and nerve fibers. Delivery of a drug from a single point source also limits the distribution of the drug by decreasing the effective radius of penetration of the drug agent into the surrounding tissue receptor population. Another aspect of this invention is therefore to overcome these inherent limitations of single point source drug delivery by devising a multi-lumen catheter with multiple drug release sources which effectively disperse therapeutic drug agents over a brain region containing receptors for the drug, or over an anatomically extensive area of brain pathology.